You are right in point, but your calculation doesn't work.
It is true that 30G currently is ridiculous.
But a lot of people send big attachments to each other - important statistical data, source code for large programs and of course funny films with monkeys.
If you just send text files, your estimations of 16M a year may work, but a large portion of us use much, much more than that.
There is a fundamental difference betwen.pdf and for example.doc, and that is the line breaks. If you open a word document with too small fonts, you can highlight all text and increase the font size as much as you like. You cannot do that in PDF.
There are admittedly a lot of other things you can do with PDF - increase the page size, use some tool for people with poor eye-sight, like Universal Access on MacOS, but as you increase it you sooner or later will come to the state where you will have to scroll horizontally to read an full line.
"Which adds the ability to natively edit PDF document in what way exactly?"
The article doesn't say anything about "editing" PDF - just "supporting" it, and that may mean anything. If they had actually meant "edit", I'm pretty sure they would have said so. The choice of the fuzzier word "support" probably means that they will be able to output PDF in some very crude format.
There is no reason to assume that this will replace any of the advance features of Adobe Acrobat or other Adobe products.
A solution that would be kinder to the competition would be to have a system wide PDF printer driver, like MacOS X has. In that way you could print to PDF from any application.
Isn't there such a thing hanging around as freeware already in Windows, btw?
"All the same, the various Indo-European languages vary greatly in grammar, and we might never have recognized the family's existence if grammar was all we looked at."
I'm not sure what you are after here. Do you claim that there are no shared grammatical features between the Indo-European languages? There may be no single rule that applies to all IE languages, but there are a lot of things that are strikingly Indo-European.
* Verb conjucation by person (I am, you are, it is...) exists in all IE languages I know of, either now, or in a recent form. (Danish, Norwegian and Swedish have all but lost the feature in the modern standard form, but it existed until less than a hundred years ago, and some dialects still use it.)
* Noun gender exists in all IE languages I know of except English, and if you go back a few hundred years, you see it in English as well.
* Adjective-noun correlation is of course linked to gender, and that is also very common (un homme heureux/une femme heureuse).
* Special forms for cases (genitive, dative, accusative and so on) exists everywhere in the IE languages, afaik, even though it may be limited in the current forms of the languages.
* Subject Verb Object (SVO) word order is by a large margin the most common one among at least the European IE languages.
You will find some of those features in some non IE languages, but hardly the full set. Of those features Chinese has only SVO. Japanese has none.
I am so disgusted by the developers and their responses to my pleas for improvement in key areas that I've stopped promoting OO to people that need a cheap office suite.
I agree that OO is far behind MSO in several areas, but I'm surprised at your using the word "disgusted" here. Those of the developers who are at Sun surely have to listen to what Sun's management says. And those who contribute to the project for idealistic reasons surely have no duty to do anything but what they fancy themselves. Clearly we have reason to be more grateful, if they do what we want, but even if they don't, it is difficult to see why we would be disgusted for not getting exactly what we want for free.
Why would you avoid wikipedia for detailed academic research? I realise you cannot quote it as is, but in about any subject I can think of, it is an excellent source of compiled facts, which one then can verify elsewhere as needed.
This may be just me, but I feel utterly frustrated that I have to carry around all these devices.
1 iPod to store data and listen to music and read non European text files.
1 Palm to visualise pdf-files and to store new documents written on the road.
1 Palm ultrathin keyboard to enter the new documents to the Palm properly and quickly.
1 Phone to... ehm... phone.
1 Camera to take pictures with a decent quality.
Add to that a Gecko GPS for hiking.
It should be possible and not too difficult to fit all of those ones in the same not clunky device. Ok, let's accept that the keyboard has to be a separate thing.
Pleeeeease, Apple, Dell, HP, Nokia, Palm, Sony Ericsson, or whoever, give me one for X-mas.
Then he wont need social skills - he can kick the bully's asses and get back to doing what he loves.
Unfortunately this is wrong in many cases. The absolutely worst programmers I have met have been very bright people with no social skills. (I realise that really dumb programmers would be even worse, but we usually don't employ those ones in the first place.)
There may be some jobs out there for people who sit in a corner and write excellent code. However, most projects involve several persons, and unless you learn to communicate properly, taking the time to express things so other people understand them and taking the time to try to understand what your less intelligent colleagues actually mean, there is a real risk that your presence is rather damaging to the project.
(And, yes, the best programmers I have met have obviously also been very bright.)
I'm surprised at that. I've used Mozilla for many years on MacOS, Windows NT and Windows 2000. As far as I remember, I have never ever had any OS crashes due to Mozilla. Could it be the combination of Mozilla and something else you installed? Or is XP really that bad?
This is a very good illustration of the attitude I describe in my original post: "Ironically, it seems a lot of the remaining programmers try to make things better than the standards, instead of following them, ignoring the fact that a new standard is better than the old one only if it actually becomes a standard."
It's faster to work in Gimp than PSP.
Which is why I in my original post wrote: "I now love Gimp."
And no, I'm not a Unix elitist or whatever you want to call me.
I prefer not to call you anything at all, neither bad nor good, as I don't know you. However, rest assured that the thought had never struck me to call you a Unix elitist.
The first time you start Gimp, it shows a Tips window telling you that most actions are done using the right mouse button.
That's exactly the kind of attitude I'm talking about. "It's not difficult, you just do like this..." "You just read the Tips window..." "You just go to the man pages.." "You just spend the rest of the weekend learning a new paradigm instead of actually saving the file, which you anyhow probably won't be able to do, as they hid the Save command in a pop-up menu, where 90% of the first time user won't look."
Let's say that the client has 10000 users, who all spend 5 minutes trying to figure out how to open a file in a "Load image dialog" that follows no standards whatsoever, and which lists texts files! (Common.txt-files are listed in an image editor's "Load image" dialog!) The users are supposed to add the text "Copyright Acme" to the picture, so they go to the Text tool, where they are supposed to write in the field which is called "Preview" (!), and which contains the letters "abcdefghijk ABCDEFGHIJK" (!!). They get to know that they have chosen a 2-byte font, which "may not be displayed correctly", but they have no idea why this font was displayed, which fonts are "good", or indeed what "may" and "correctly" mean. As it is an official image, they spend at least 30 minutes trying to make sure the font they want is good for printing and distribution. And then they cannot save.
Altogether it can take roughly one hour for the average user to open his first impage in Gimp, add a textstring to it and save and close the image.
10000 users times 1 hour gives about five man years lost to bad usability. For simple actions. In one application.
Yes, it can be done in no more than two minutes, once you know how to use the application. But if you write software only for people, who are willing to spend time on learning and who are smart enough to look in the right places, then you have limited the potential customer group to a fraction of the population.
The problem with Linux usability is that half of the programmers write software primarily for themselves, so from their perspective, it does not need to be usable for everyone. The incentive is simply not there. And it is understable that they prefer doing other things. Writing programs with good usability is expensive.
Ironically, it seems a lot of the remaining programmers try to make things better than the standards, instead of following them, ignoring the fact that a new standard is better than the old one, only if it actually becomes a standard.
Gimp, which is discussed elsewhere here, does basically no single thing right. The "application" itself (at least in version 1.2.3) has a File menu but it does not have the standard "Print" or "Close (file)" commands, that we expect in Windows or MacOS. There is no Edit menu. There is no Windows menu and the "Help" does not contain any search function.
Don't get me wrong. I now love Gimp. I use it regularly. But when I first saw it, I had no idea even on where to start to explore the application. Gimp imposed an additional learning process on me, which I could have done without.
Another favourite example of bad UI is Scilab (version 2.7 in my case). It is an application without any menus at all - at first sight. It turns out that the "File" menu is a button on the page. It doesn't contain the standard items of course. And there is no Edit button or menu. In the File "menu" there is a "Kill" option, which kills the application without any confirmation message, so if you used it thinking that something in the "File" "menu" would kill the current file only, you lost all your unsaved data. And so on.
I know where to look for certain things, and I want them to be where they are in other applications.
At least since 1984 it has been known in the computer world that usability standards were important. The linux/unix/x world has still not fully grasped it 20 years later. Hopefully experiences like Munich will improve the situation.
Slashdot Mirror
It is true that 30G currently is ridiculous.
But a lot of people send big attachments to each other - important statistical data, source code for large programs and of course funny films with monkeys.
If you just send text files, your estimations of 16M a year may work, but a large portion of us use much, much more than that.
There are admittedly a lot of other things you can do with PDF - increase the page size, use some tool for people with poor eye-sight, like Universal Access on MacOS, but as you increase it you sooner or later will come to the state where you will have to scroll horizontally to read an full line.
The article doesn't say anything about "editing" PDF - just "supporting" it, and that may mean anything. If they had actually meant "edit", I'm pretty sure they would have said so. The choice of the fuzzier word "support" probably means that they will be able to output PDF in some very crude format.
There is no reason to assume that this will replace any of the advance features of Adobe Acrobat or other Adobe products.
Isn't there such a thing hanging around as freeware already in Windows, btw?
I'm not sure what you are after here. Do you claim that there are no shared grammatical features between the Indo-European languages? There may be no single rule that applies to all IE languages, but there are a lot of things that are strikingly Indo-European.
* Verb conjucation by person (I am, you are, it is...) exists in all IE languages I know of, either now, or in a recent form. (Danish, Norwegian and Swedish have all but lost the feature in the modern standard form, but it existed until less than a hundred years ago, and some dialects still use it.)
* Noun gender exists in all IE languages I know of except English, and if you go back a few hundred years, you see it in English as well.
* Adjective-noun correlation is of course linked to gender, and that is also very common (un homme heureux/une femme heureuse).
* Special forms for cases (genitive, dative, accusative and so on) exists everywhere in the IE languages, afaik, even though it may be limited in the current forms of the languages.
* Subject Verb Object (SVO) word order is by a large margin the most common one among at least the European IE languages.
You will find some of those features in some non IE languages, but hardly the full set. Of those features Chinese has only SVO. Japanese has none.
I agree that OO is far behind MSO in several areas, but I'm surprised at your using the word "disgusted" here. Those of the developers who are at Sun surely have to listen to what Sun's management says. And those who contribute to the project for idealistic reasons surely have no duty to do anything but what they fancy themselves. Clearly we have reason to be more grateful, if they do what we want, but even if they don't, it is difficult to see why we would be disgusted for not getting exactly what we want for free.
Why would you avoid wikipedia for detailed academic research? I realise you cannot quote it as is, but in about any subject I can think of, it is an excellent source of compiled facts, which one then can verify elsewhere as needed.
1 iPod to store data and listen to music and read non European text files.
1 Palm to visualise pdf-files and to store new documents written on the road.
1 Palm ultrathin keyboard to enter the new documents to the Palm properly and quickly.
1 Phone to... ehm... phone.
1 Camera to take pictures with a decent quality.
Add to that a Gecko GPS for hiking.
It should be possible and not too difficult to fit all of those ones in the same not clunky device. Ok, let's accept that the keyboard has to be a separate thing.
Pleeeeease, Apple, Dell, HP, Nokia, Palm, Sony Ericsson, or whoever, give me one for X-mas.
Unfortunately this is wrong in many cases. The absolutely worst programmers I have met have been very bright people with no social skills. (I realise that really dumb programmers would be even worse, but we usually don't employ those ones in the first place.)
There may be some jobs out there for people who sit in a corner and write excellent code. However, most projects involve several persons, and unless you learn to communicate properly, taking the time to express things so other people understand them and taking the time to try to understand what your less intelligent colleagues actually mean, there is a real risk that your presence is rather damaging to the project.
(And, yes, the best programmers I have met have obviously also been very bright.)
I'm surprised at that. I've used Mozilla for many years on MacOS, Windows NT and Windows 2000. As far as I remember, I have never ever had any OS crashes due to Mozilla. Could it be the combination of Mozilla and something else you installed? Or is XP really that bad?
This is a very good illustration of the attitude I describe in my original post: "Ironically, it seems a lot of the remaining programmers try to make things better than the standards, instead of following them, ignoring the fact that a new standard is better than the old one only if it actually becomes a standard."
It's faster to work in Gimp than PSP.
Which is why I in my original post wrote: "I now love Gimp."
And no, I'm not a Unix elitist or whatever you want to call me.
I prefer not to call you anything at all, neither bad nor good, as I don't know you. However, rest assured that the thought had never struck me to call you a Unix elitist.
That's exactly the kind of attitude I'm talking about. "It's not difficult, you just do like this..." "You just read the Tips window..." "You just go to the man pages.." "You just spend the rest of the weekend learning a new paradigm instead of actually saving the file, which you anyhow probably won't be able to do, as they hid the Save command in a pop-up menu, where 90% of the first time user won't look."
Let's say that the client has 10000 users, who all spend 5 minutes trying to figure out how to open a file in a "Load image dialog" that follows no standards whatsoever, and which lists texts files! (Common .txt-files are listed in an image editor's "Load image" dialog!) The users are supposed to add the text "Copyright Acme" to the picture, so they go to the Text tool, where they are supposed to write in the field which is called "Preview" (!), and which contains the letters "abcdefghijk ABCDEFGHIJK" (!!). They get to know that they have chosen a 2-byte font, which "may not be displayed correctly", but they have no idea why this font was displayed, which fonts are "good", or indeed what "may" and "correctly" mean. As it is an official image, they spend at least 30 minutes trying to make sure the font they want is good for printing and distribution. And then they cannot save.
Altogether it can take roughly one hour for the average user to open his first impage in Gimp, add a textstring to it and save and close the image.
10000 users times 1 hour gives about five man years lost to bad usability. For simple actions. In one application.
Yes, it can be done in no more than two minutes, once you know how to use the application. But if you write software only for people, who are willing to spend time on learning and who are smart enough to look in the right places, then you have limited the potential customer group to a fraction of the population.
Ironically, it seems a lot of the remaining programmers try to make things better than the standards, instead of following them, ignoring the fact that a new standard is better than the old one, only if it actually becomes a standard.
Gimp, which is discussed elsewhere here, does basically no single thing right. The "application" itself (at least in version 1.2.3) has a File menu but it does not have the standard "Print" or "Close (file)" commands, that we expect in Windows or MacOS. There is no Edit menu. There is no Windows menu and the "Help" does not contain any search function.
Don't get me wrong. I now love Gimp. I use it regularly. But when I first saw it, I had no idea even on where to start to explore the application. Gimp imposed an additional learning process on me, which I could have done without.
Another favourite example of bad UI is Scilab (version 2.7 in my case). It is an application without any menus at all - at first sight. It turns out that the "File" menu is a button on the page. It doesn't contain the standard items of course. And there is no Edit button or menu. In the File "menu" there is a "Kill" option, which kills the application without any confirmation message, so if you used it thinking that something in the "File" "menu" would kill the current file only, you lost all your unsaved data. And so on.
I know where to look for certain things, and I want them to be where they are in other applications.
At least since 1984 it has been known in the computer world that usability standards were important. The linux/unix/x world has still not fully grasped it 20 years later. Hopefully experiences like Munich will improve the situation.