What if your running GNU/Solaris? The GNU speaks to specific things, just as Solaris or Linux speaks to specific things. What would be fun, though, would be to call it GNU/Unix...;-)
I'll take this a step further. Science is a useful methodology for the building of models that are extremely useful for making specific predictions. Whenever you forget that its a model, and start BELIEVING it, its no longer science. Most laypeople don't really care about science, or models, so long as "it just works" when you flip the switch. However, they like certitude. It makes them comfortable that when flipping the switch next time, the TV will keep working. Now, Newtonian Mechanics, for instance, works pretty damn well for building bridges. Is it real? Nope. But its a really useful model. It works within a context (things are large enough we can ignore quantum effects, relative velocities are low enough that relativity isn't worth considering, etc...). To say that Newtonian Mechanics is "right" or "wrong" misses not only the mark, but the target. Science is never "right" nor "wrong" in the manner in which you used the terms. It is, however, extremely useful.
I think I sense a chicken/egg sort of paradox here. "Entropy is just the degree to which a system has moved from a higher to a lover potential." So you have defined entropy in terms of energy changes. Likewise, can't potential energy itself be defined (that is, the structured state from which said potential arises) in terms of entropy? Does a potential actualizing lead to entropy, or does the initial state (divided) and final state (difference "dispersed") merely describe an entropic process, with potentials being a manner of describing order yet to be dissipated? I dare suggest energy isn't more fundamental than entropy.
Try this. You have a well organized study. Books are organized on shelves. Pencils and pens and blank paper are laid out on a desk. But no one is doing maid service, and over time wind blows in and disperses the neat stack of paper. An earthquake knocks over the books. If you come in and "haul out the garbage", it might look cleaner, but now ask, "how much work do i have to do to get back to the organized study mode?", which has become *much* harder since the volumes of Britannica are now missing.
I might agree that in your analogy you are indeed creating a highly ordered end product. Unfortunately, the end product of Hawking Radiation is the exact opposite. You input highly structured, dissimilar material objects, and the result is uniformly dispersed, homogenized radiation. The differences are dissipated away. A universe of electrons, protons, and photons can result, when distributed in such a way that energy is pumped into a region, in emergent patterns in said region. Things like us, for instance. When everything is dissipated to uniform radiation, there is neither a gradient to drive, nor anything to be driven thereby. I'd want to say there is then nothing but background radiation, but I'm lost in terms of what background means without a foreground. At the limit, I'm not sure "time" itself would have meaning, though I assume that all through the approach it would just be losing, and never quite totally lose, meaning.
Also note: you had to put energy into your system to "magicly" convert the carbon into a diamond structure. Likewise, when I consume lunch, I'm feeding to maintain my equilibrium. Without that orderly, structured input of usable energy, I'd dissipate and disperse. If we had an external source of energy, we could perhaps likewise impose order on the Hawking Radiation and convert it into useful structures such as electrons and protons. Without such an external source, though, whatever that means, the future is (eventually) bleak.
The standards aren't broken. What is broken is Microsoft's browsers. Its easier to build plugins to fix IE than it is to wrestle control of Microsoft away and force them to fix IE. Hopefully, if this plugin sees enough use, that in itself will leverage MS to fix IE. Note: it will NOT be the same mess as today, because instead of having to code for Gecko, Trident, and Webkit, you could choose one (preferably standards compliant) and code to just that one, and let the plugin do the work. While the *easy* way would be for IE to become standards compliant and make this unnecessary that just doesn't seem to be in the cards.
Well, if your asking for a vote, I'd vote to give W3C some teeth. Seems to me it should be, 1) User of the browser. 2) W3C. 3) Browser. 4) Web site author.
Nice to see their words in context. It still doesn't make sense to me, though. Rather than fragmenting the web, I see this as a way to unify the web behind standards. This is a stepping stone for most of the browsing world to get free, and how can that not be good? If we end up, three years from now, in a world where either IE renders correctly, or nobody surfs w/o the Chrome Plugin, thats also a very good thing.
Spoken like an employed fool. From a developer who is employed, if I can code to standards and use the existence of this plugin to ensure that the majority of browsers will now work with standards, its a win-win for me and my customers. Soon, every major web site will support HTML5 and CSS3, and life will be good. You are just a dinosaur. Catch up or be left behind.
No, thats just totally wrong. Instead of yet another spec to code for, now you can code to the standard, and expect most people to be able to view it correctly. Why code for IE6, IE7, IE8, and "all the rest of the world"? Code HTMl5 and CSS3, and let it just work the way its supposed to. This is, I think, what is actually scaring MS. Google's plugin makes it possible for the majority of the web to ignore MS specs and use standards, transparently. You know how developers are going to respond to this. Its like Christmas and Thanksgiving all at the same time.
you "took it"...who did you "take" it from such that they can't use it now? oh, so they still have it? then you didn't really take it, you COPIED it....
I think what he meant by relevance of users is that just because IE6 has almost 15% market share, that doesn't mean that 15% of your customers use IE6. Large corporations are tied to IE6 so its used at work, during the week, but its share goes down on weekends when people surf from home. If it so happens that your web site isn't visited much by people at work, say, but rather caters to people surfing from home, then it might be that IE6 is less than 1/2 of a percentage point of your customers. From a purely financial perspective if it costs more to support them that they bring in in revenue, then they aren't really customers. They are parasites. At least, thats how I interpreted his relevance remark.
"...because I have an issue with the way MS does business..."
While a good case can be made regarding the way Microsoft goes about doing business, the fact of the matter is really that they can't ever seem to be standards compliant. They are always playing catch up, and what works isn't linear. Different things break in different places as you move up the IE version numbers. This isn't about them being thieves (as you seem to think), its about the lack of competence displayed by IE. Basicly it comes down to coding for 4 main browser groups: IE6, IE7, (soon to be IE8), and the rest of the web.
Its amazing to see someone refering to IE6 and suggesting that someone other than microsoft did something wrong. Wake up, face reality, and realize where the "too lazy" to do it right tag really belongs. "Take more pride in their work." In context, thats rich!
You draw the line, sure. But does it do what you intend? Is the data transfer synchronous or asynchronous? Do you need a buffer? Soft real-time? Hard? It just doesn't matter? Are going to want events triggered if the input signal reaches certain stages? Does the time rate of change of the signal matter as well as the value? Oh? You want a weighted average of the value and its rate of change?
Don't get me wrong. Its easy to design a system like you describe so long as you like what I feel that line represents. Hell on you if you don't, though. If you want input as to what that line represents, you'll have to express your intentions. We use language for that.
What you describe is a valid reason: sticking with the status que in a team environment where the team has already chosen tools.
Throwing away a set of tools, however, because he *doesn't* want to use an open-source one, without further qualification, is a bias.
The difference is that you explain why moving to a different tool chain would be hard. He presented as a reason merely that it was open-source. See the difference?
Its one thing to hire Sally instead of Fred because Sally speaks 4 languages (lets stipulate these languages matter for this job description). Its another thing to hire Sally because he *doesn't* want to hire a guy.
As much as it pains me, I have to admit that substituting MS Word print-to-PDF for a LaTeX-to-PDF via LyX (or Kile, or even Emacs,) is certainly fair (enough) and within the scope of the methodology I'm exposing. Said methodology being that
PDF be treated as an end product format, with another editable format that can be easily transformed to PDF.
So then why don't you do that? Are there other criteria of which I'm unaware? Do you need to edit PDFs that were generated by third parties? Are they internal to your organization, and would it be possible to implement a policy of "no PDF w/o [fill in your editable format] source included"?
Just a little googling dug up the question of PDF-to-LaTeX conversion in the Ubuntu forums, which would rather sweetly allow round-trip reverse engineering.
Seems less than encouraging:
Re: Pdf to Latex convesrion
remember that PDF's main design goal is visual layout. it does not contain structural information.
a latex will specify that something is a section or subsection heading, pdf only says what font,size and position the words should have.
also latex keeps track of figures, tables and references, and only puts the numbers in when exporting to pdf (or dvi if you are oldschool). the pdf file will just have the numbers, not the associations.
you would need clever code to rebuild all the information that is lost in a pdf file.
Looks to me like we're back to the central difference between document structure and document presentation once again. Structure is amendable to algorithmic processing. Presentation in its purest form (think a jpeg of a handwritten note), while perhaps giving pixel by pixel control of display, is much less amendable to, for instance, sweeping regex modifications. Hence the desire to keep the two separate, and my suggestion to not let go of the "source" when you've built the "binary", so to speak.
First, its true that the average user would get better results with less training using LyX, without ever having to learn LaTeX.
Second, it seems to me now that you are confusing formating with presentation. You do know, for instance, applying emphasis is formating, but whether emphasis means to display in bold or italics is not formating, but presentation? Likewise, you format your document by indicating the location of paragraphs and lists. How this is presented, though, isn't formating. Does a paragraph start with an indentation? Not formating, thats what a style sheet determines. Do you use one or two or three columns? You don't use columns in formating. You denote the structural information, and let your program flow the formated data into however many columns your chosen style sheet uses. This also means that changing margins or number of columns is simple. You change the style sheet ad recompile. Again, formating has merely to do with denoting the structural characteristic of a section, not its appearance.
The reason that LyX works so well for the untrained newbies is that the newbie selects a section type from a drop down menu, but the presentation of that characteristic is done via style sheets written by experts. Thus, a newbie can concentrate on what they want to say, while LyX insures it looks like it were professionally typeset. You can't get that using MS Word (as a newbie) unless you hire a professional to apply formating for you. You can't get it that easily, even if you are said professional, using MS Word.
Font selection is an interesting thing to bring up. Font selection isn't part of formating, except as the style sheet would apply different fonts (and font sizes, etc...), to different structures in the document. It sounds to me like what you mean by fine-grained control is to throw away all structural consistency and rather than format a document, do formating and presentation word by word by word. That is insane. Its also not allowed in the professional journals with which I am acquainted.
Pagination is only one small part of formating, and its demise is only a small part of the author's argument. The author was saying, I believe, that continued pagination is stupid when and where there is no longer an inherent need for pagination.
I really wouldn't consider Word "fine-grained", by the way. If you need fine-grained control, use LaTeX, if you need ultra-fine grained control, there is TeX. Its about the best there is.
Actually I almost never print LaTeX. I use it to generate PDF or MathML. Gnumeric exports its spreadsheets to LaTeX, too. I admit that Office can do some of what LaTeX does, but right after I switched I had people approach me to ask how I had made my reports look so good. It looked like commercial printing, and they wondered what it cost to outsource it to get level of quality. It really looks that much better.
I don't create PDF directly. I create LaTeX, then covert it to PDF. So I've not noticed this problem of which you speak, but then again, I'm not trying to modify PDF directly, rather I'm producing a PDF with necessary changes.
It'll help if you think of LaTeX as the source code, and PDF as the compiled binary. Then its no surprise that direct, sophisticated editing isn't easy. (Likewise, it sure wasn't easy back in the day, using debug to directly edit a binary file). However, making changes to the source *is* easy. Using LyX its laid out in a sweet GUI. It does word wrap for me. Then I just "recompile" (actually, export) the LaTeX into PDF.
"He keeps it because he still *needs* to format documents,"
as do I
and *wants* to do so using a WYSIWIG editor --
and I use a WYSIWYM editor, which seems to have every advantage of WYSIWYG, so there is no functional loss, and a lot to be gained
and *doesn't* want to (or can't, maybe, due to company policy?) use an open-source one.
which is the most honest reason I've heard yet: irrational bias.
(Now don't get me wrong, I respect irrational bias. Not so much for myself, I mean, but it is especially good when employed by those I'm competing with.)
Slashdot Mirror
What if your running GNU/Solaris? The GNU speaks to specific things, just as Solaris or Linux speaks to specific things. What would be fun, though, would be to call it GNU/Unix... ;-)
I'll take this a step further. Science is a useful methodology for the building of models that are extremely useful for making specific predictions. Whenever you forget that its a model, and start BELIEVING it, its no longer science. Most laypeople don't really care about science, or models, so long as "it just works" when you flip the switch. However, they like certitude. It makes them comfortable that when flipping the switch next time, the TV will keep working. Now, Newtonian Mechanics, for instance, works pretty damn well for building bridges. Is it real? Nope. But its a really useful model. It works within a context (things are large enough we can ignore quantum effects, relative velocities are low enough that relativity isn't worth considering, etc...). To say that Newtonian Mechanics is "right" or "wrong" misses not only the mark, but the target. Science is never "right" nor "wrong" in the manner in which you used the terms. It is, however, extremely useful.
I think I sense a chicken/egg sort of paradox here. "Entropy is just the degree to which a system has moved from a higher to a lover potential." So you have defined entropy in terms of energy changes. Likewise, can't potential energy itself be defined (that is, the structured state from which said potential arises) in terms of entropy? Does a potential actualizing lead to entropy, or does the initial state (divided) and final state (difference "dispersed") merely describe an entropic process, with potentials being a manner of describing order yet to be dissipated? I dare suggest energy isn't more fundamental than entropy.
Try this. You have a well organized study. Books are organized on shelves. Pencils and pens and blank paper are laid out on a desk. But no one is doing maid service, and over time wind blows in and disperses the neat stack of paper. An earthquake knocks over the books. If you come in and "haul out the garbage", it might look cleaner, but now ask, "how much work do i have to do to get back to the organized study mode?", which has become *much* harder since the volumes of Britannica are now missing.
I might agree that in your analogy you are indeed creating a highly ordered end product. Unfortunately, the end product of Hawking Radiation is the exact opposite. You input highly structured, dissimilar material objects, and the result is uniformly dispersed, homogenized radiation. The differences are dissipated away. A universe of electrons, protons, and photons can result, when distributed in such a way that energy is pumped into a region, in emergent patterns in said region. Things like us, for instance. When everything is dissipated to uniform radiation, there is neither a gradient to drive, nor anything to be driven thereby. I'd want to say there is then nothing but background radiation, but I'm lost in terms of what background means without a foreground. At the limit, I'm not sure "time" itself would have meaning, though I assume that all through the approach it would just be losing, and never quite totally lose, meaning.
Also note: you had to put energy into your system to "magicly" convert the carbon into a diamond structure. Likewise, when I consume lunch, I'm feeding to maintain my equilibrium. Without that orderly, structured input of usable energy, I'd dissipate and disperse. If we had an external source of energy, we could perhaps likewise impose order on the Hawking Radiation and convert it into useful structures such as electrons and protons. Without such an external source, though, whatever that means, the future is (eventually) bleak.
The standards aren't broken. What is broken is Microsoft's browsers. Its easier to build plugins to fix IE than it is to wrestle control of Microsoft away and force them to fix IE. Hopefully, if this plugin sees enough use, that in itself will leverage MS to fix IE. Note: it will NOT be the same mess as today, because instead of having to code for Gecko, Trident, and Webkit, you could choose one (preferably standards compliant) and code to just that one, and let the plugin do the work. While the *easy* way would be for IE to become standards compliant and make this unnecessary that just doesn't seem to be in the cards.
Well, if your asking for a vote, I'd vote to give W3C some teeth. Seems to me it should be, 1) User of the browser. 2) W3C. 3) Browser. 4) Web site author.
Nice to see their words in context. It still doesn't make sense to me, though. Rather than fragmenting the web, I see this as a way to unify the web behind standards. This is a stepping stone for most of the browsing world to get free, and how can that not be good? If we end up, three years from now, in a world where either IE renders correctly, or nobody surfs w/o the Chrome Plugin, thats also a very good thing.
Spoken like an employed fool. From a developer who is employed, if I can code to standards and use the existence of this plugin to ensure that the majority of browsers will now work with standards, its a win-win for me and my customers. Soon, every major web site will support HTML5 and CSS3, and life will be good. You are just a dinosaur. Catch up or be left behind.
No, thats just totally wrong. Instead of yet another spec to code for, now you can code to the standard, and expect most people to be able to view it correctly. Why code for IE6, IE7, IE8, and "all the rest of the world"? Code HTMl5 and CSS3, and let it just work the way its supposed to. This is, I think, what is actually scaring MS. Google's plugin makes it possible for the majority of the web to ignore MS specs and use standards, transparently. You know how developers are going to respond to this. Its like Christmas and Thanksgiving all at the same time.
you "took it"...who did you "take" it from such that they can't use it now? oh, so they still have it? then you didn't really take it, you COPIED it....
..."and don't have to worry about forwards compatibility to Firefox", there, fixed that for you.
In what reality does 15% equate to "most common tool". That and the "also-ran" indicate you are trolling.
I think what he meant by relevance of users is that just because IE6 has almost 15% market share, that doesn't mean that 15% of your customers use IE6. Large corporations are tied to IE6 so its used at work, during the week, but its share goes down on weekends when people surf from home. If it so happens that your web site isn't visited much by people at work, say, but rather caters to people surfing from home, then it might be that IE6 is less than 1/2 of a percentage point of your customers. From a purely financial perspective if it costs more to support them that they bring in in revenue, then they aren't really customers. They are parasites. At least, thats how I interpreted his relevance remark.
"...because I have an issue with the way MS does business..."
While a good case can be made regarding the way Microsoft goes about doing business, the fact of the matter is really that they can't ever seem to be standards compliant. They are always playing catch up, and what works isn't linear. Different things break in different places as you move up the IE version numbers. This isn't about them being thieves (as you seem to think), its about the lack of competence displayed by IE. Basicly it comes down to coding for 4 main browser groups: IE6, IE7, (soon to be IE8), and the rest of the web.
I always thought that Adobe implemented a poor SVG to prevent SVG competing with Flash.
Its amazing to see someone refering to IE6 and suggesting that someone other than microsoft did something wrong. Wake up, face reality, and realize where the "too lazy" to do it right tag really belongs. "Take more pride in their work." In context, thats rich!
You draw the line, sure. But does it do what you intend? Is the data transfer synchronous or asynchronous? Do you need a buffer? Soft real-time? Hard? It just doesn't matter? Are going to want events triggered if the input signal reaches certain stages? Does the time rate of change of the signal matter as well as the value? Oh? You want a weighted average of the value and its rate of change?
Don't get me wrong. Its easy to design a system like you describe so long as you like what I feel that line represents. Hell on you if you don't, though. If you want input as to what that line represents, you'll have to express your intentions. We use language for that.
What you describe is a valid reason: sticking with the status que in a team environment where the team has already chosen tools.
Throwing away a set of tools, however, because he *doesn't* want to use an open-source one, without further qualification, is a bias.
The difference is that you explain why moving to a different tool chain would be hard. He presented as a reason merely that it was open-source. See the difference?
Its one thing to hire Sally instead of Fred because Sally speaks 4 languages (lets stipulate these languages matter for this job description). Its another thing to hire Sally because he *doesn't* want to hire a guy.
PDF be treated as an end product format, with another editable format that can be easily transformed to PDF .
So then why don't you do that? Are there other criteria of which I'm unaware? Do you need to edit PDFs that were generated by third parties? Are they internal to your organization, and would it be possible to implement a policy of "no PDF w/o [fill in your editable format] source included"?
Just a little googling dug up the question of PDF-to-LaTeX conversion in the Ubuntu forums, which would rather sweetly allow round-trip reverse engineering.
Seems less than encouraging:
Looks to me like we're back to the central difference between document structure and document presentation once again. Structure is amendable to algorithmic processing. Presentation in its purest form (think a jpeg of a handwritten note), while perhaps giving pixel by pixel control of display, is much less amendable to, for instance, sweeping regex modifications. Hence the desire to keep the two separate, and my suggestion to not let go of the "source" when you've built the "binary", so to speak.
First, its true that the average user would get better results with less training using LyX, without ever having to learn LaTeX.
Second, it seems to me now that you are confusing formating with presentation. You do know, for instance, applying emphasis is formating, but whether emphasis means to display in bold or italics is not formating, but presentation? Likewise, you format your document by indicating the location of paragraphs and lists. How this is presented, though, isn't formating. Does a paragraph start with an indentation? Not formating, thats what a style sheet determines. Do you use one or two or three columns? You don't use columns in formating. You denote the structural information, and let your program flow the formated data into however many columns your chosen style sheet uses. This also means that changing margins or number of columns is simple. You change the style sheet ad recompile. Again, formating has merely to do with denoting the structural characteristic of a section, not its appearance.
The reason that LyX works so well for the untrained newbies is that the newbie selects a section type from a drop down menu, but the presentation of that characteristic is done via style sheets written by experts. Thus, a newbie can concentrate on what they want to say, while LyX insures it looks like it were professionally typeset. You can't get that using MS Word (as a newbie) unless you hire a professional to apply formating for you. You can't get it that easily, even if you are said professional, using MS Word.
Font selection is an interesting thing to bring up. Font selection isn't part of formating, except as the style sheet would apply different fonts (and font sizes, etc...), to different structures in the document. It sounds to me like what you mean by fine-grained control is to throw away all structural consistency and rather than format a document, do formating and presentation word by word by word. That is insane. Its also not allowed in the professional journals with which I am acquainted.
Pagination is only one small part of formating, and its demise is only a small part of the author's argument. The author was saying, I believe, that continued pagination is stupid when and where there is no longer an inherent need for pagination.
I really wouldn't consider Word "fine-grained", by the way. If you need fine-grained control, use LaTeX, if you need ultra-fine grained control, there is TeX. Its about the best there is.
Actually I almost never print LaTeX. I use it to generate PDF or MathML. Gnumeric exports its spreadsheets to LaTeX, too. I admit that Office can do some of what LaTeX does, but right after I switched I had people approach me to ask how I had made my reports look so good. It looked like commercial printing, and they wondered what it cost to outsource it to get level of quality. It really looks that much better.
I don't create PDF directly. I create LaTeX, then covert it to PDF. So I've not noticed this problem of which you speak, but then again, I'm not trying to modify PDF directly, rather I'm producing a PDF with necessary changes.
It'll help if you think of LaTeX as the source code, and PDF as the compiled binary. Then its no surprise that direct, sophisticated editing isn't easy. (Likewise, it sure wasn't easy back in the day, using debug to directly edit a binary file). However, making changes to the source *is* easy. Using LyX its laid out in a sweet GUI. It does word wrap for me. Then I just "recompile" (actually, export) the LaTeX into PDF.
"He keeps it because he still *needs* to format documents,"
as do I
and *wants* to do so using a WYSIWIG editor --
and I use a WYSIWYM editor, which seems to have every advantage of WYSIWYG, so there is no functional loss, and a lot to be gained
and *doesn't* want to (or can't, maybe, due to company policy?) use an open-source one.
which is the most honest reason I've heard yet: irrational bias.
(Now don't get me wrong, I respect irrational bias. Not so much for myself, I mean, but it is especially good when employed by those I'm competing with.)