1. As others have said, "The Elements of Style". In particular the rule "omit needless words".
2. Ignore advice to prefer the active voice. The passive voice is appropriate for much techical writing. "The modulation was measured in the standard CCIR bandwidth" is appropriate. "I measured the modulation..." is chatty and unnecessary. The method and the results are important; the identity of the experimenter is usually not.
3. Never write anything you cannot substantiate. Know what you would cite as a source if you were asked.
4. Don't bullshit. If you find yourself using vague words such as "significantly", "mostly", "highly", etc., replace them with actual numbers. If you cannot then either delete the sentence, or go do some more research until you can.
5. Structure. Write the outline first and fill in the details later. Do not be afraid to reorganise if it becomes clear that a different structure would be clearer or simpler.
I think the market for number of people needing > 64K rows must be pretty slim.
Excel is very handy for ad hoc analysis of data. It is common for me to leave a test or an experiment running overnight, logging data to a text file, then import the results into excel to get a quick look at what I have.
The 64k rows limit is tedious, and even worse is that the number of points in a plot seems to be limited to 32k. If you plot more it simply fails silently and omits the missing rows. (I generally use x-y scatter plots, perhaps there is a limit of 64k values, each pair counting as two).
These days I use gnuplot to quickly plot these kind of ad hoc graphs, but it would be convenient in many situations if spreadsheets would handle unlimited rows.
Everyone in the country has access to phone service, no matter how rural / remote they are. This has been a tremendously important program...
Important to whom? Oh, the people who choose to live in the middle of nowhere. Perhaps they should pay for it, rather than expecting their choice to be subsidized by the rest of us.
what's the rule for determining what needs to be clicked and what needs to be double-clicked?
The rule is: icons that are metaphores for things (like folders or documents) are single-clicked to select and double-clicked to activate; icons that are metaphores for controls (like command buttons or menu items or just about any other widget) are single-clicked to operate the control.
Re:Now -this- is the stuff of nightmares.
on
HP Buys Compaq
·
· Score: 1
I just hope HP sells of the stuff they make that doesn't suck (calculators, printers, and medical/testing equipment) before they make something really stupid and tank.
Re:TiVo is actually *good* for advertisors
on
Calling Out TiVo
·
· Score: 1
Let's... assume that we *want* network television to continue to be free.
I do not accept this premise because it is not true. Network TV is not free. It simply has no cost at the point of delivery. Which, as anyone who has just done their taxes will tell you, is not the same as free.
I'd prefer to skip commercials altogether, but the economic realities mean someone has to pay... I'd much rather pay... by watching some commercials targeted at me specifically than paying [money]
You do pay real money for ad-supported TV. You pay through the increased prices of goods and services that are advertised. When you buy a box of soap poweder in the supermarket a large part of the money you hand over goes to pay for all that crap on Fox. The money that pays for ad-supported TV does not magic itself out of thin air - it comes out of your pocket ultimately.
Although TV detector vans are the most obvious way of catching people, the real work is done by computers. They have a bloody big database containing every street address in the UK, and they subtract from this the data from licenses sold. The remainder are households that don't have a license. Since TV ownership is almost universal, almost all of these households will be license evaders.
Corporations need to motivate people to buy their products.... Capitalism requires product awareness in consumers to work at all.
I buy a new car about once every five years. During prime time TV I am presented with ads for new cars about once every eight minutes. That kind of disparity very quickly becomes annoying.
It all depends how you make your PDFs. The bulk of the size of a typical PDF is in the images, and the images may be embedded within the PDF in various formats and with various levels of compression. A comparison without detailing what format was used is meaningless.
For example, if you scan a typical one page black and white text document at 300dpi and save the result as a TIFF with CCITT Group 4 compression it will be 20-50kB. If you make a PDF by printing to Postscript then using the PDF generation tool within the GhostScript Viewer you will end up with a PDF that is several MB. That's because embedded within the PDF will be an uncompressed bitmap. If you make the PDF using the c42pdf utility your PDF will be the same size as the TIFF, 20-50kB. That's because embedded within the PDF will be the compressed bitmap.
If you use the Adobe Acrobat tools to produce your PDFs, you will find lots of options for controlling how embedded images are compressed, what the resolution is and what formats are used. You can select different options for B&W, greyscale and colour images.
I suspect the DjVu comparison is not comparing like with like. And in any case, as has been noted, anyone who generates PDFs of documents like annual reports with a scanner is missing the point.
I started to get some pain in my right hand from mousing, so I switched to using the left. I was pretty clumsy the first week, a lot better the second, and after that I no longer noticed. Nowadays I switch back and forth every few weeks to even up the wear.
So I would say it is quite possible to use the 'wrong' hand - mousing is a fairly simple motor skill which it seems one can learn pretty fast.
Think about it. The book was probably not written in HTML, most likely it was written and edited in Word or some other Word processing package. The planning would come in by determining how they want to post it back online in HTML format.
Who says HTML is the answer? If you want an online, threaded, linked version why not refer to the original slashdot articles.
A more appropriate format for a book might be PDF. The Yale Web Style Guide was a reasonable example of how this can work (before they took it away to encourage you to buy the paper copy). Whatever tool Hellmouth was authored with it could almost certainly be rendered as PDF with very little effort.
As I remember it, much of the story was actually about how expensive and less-than-ideal Irridium service was going to be due to the political deals that had to be made with third world governments (and their telcos) to get WARC to allocate them the frequencies.
Isn't this the job of Red Hat, et al? Commercial companies who seek to make money from Linux claim that what they bring to the party - that new users can't get from just downloading the sources - is support. Surely, good old market forces will deliver what new users want and need from these companies.
Slashdot Mirror
Really? I thought they were too big. Especially Jason Porritt's entry which otherwise looked good, but was far too big and would not resize.
1. As others have said, "The Elements of Style". In particular the rule "omit needless words".
2. Ignore advice to prefer the active voice. The passive voice is appropriate for much techical writing. "The modulation was measured in the standard CCIR bandwidth" is appropriate. "I measured the modulation..." is chatty and unnecessary. The method and the results are important; the identity of the experimenter is usually not.
3. Never write anything you cannot substantiate. Know what you would cite as a source if you were asked.
4. Don't bullshit. If you find yourself using vague words such as "significantly", "mostly", "highly", etc., replace them with actual numbers. If you cannot then either delete the sentence, or go do some more research until you can.
5. Structure. Write the outline first and fill in the details later. Do not be afraid to reorganise if it becomes clear that a different structure would be clearer or simpler.
Manufacturers should build this into the hardware or the bios.
They do, apparently. The press release says IBM/Lenovo ThinkPads have it in the BIOS. It can survive image reloads and hard drive swaps.
Excel is very handy for ad hoc analysis of data. It is common for me to leave a test or an experiment running overnight, logging data to a text file, then import the results into excel to get a quick look at what I have.
The 64k rows limit is tedious, and even worse is that the number of points in a plot seems to be limited to 32k. If you plot more it simply fails silently and omits the missing rows. (I generally use x-y scatter plots, perhaps there is a limit of 64k values, each pair counting as two).
These days I use gnuplot to quickly plot these kind of ad hoc graphs, but it would be convenient in many situations if spreadsheets would handle unlimited rows.
My favorite example of this is a manufacturer of programmable microwave signal generators, www.aprilinstrument.com
Ideas are not property, or at least should not be.
Important to whom? Oh, the people who choose to live in the middle of nowhere. Perhaps they should pay for it, rather than expecting their choice to be subsidized by the rest of us.
Are there any small companies that MS has had dealings with that they haven't royally screwed over?
The Hotmail guys did fairly well. The original small group of owners walked away with circa $700m IIRC.
No, it's silicon, as in sand. This is silicone.
That would be Draconian, after Draco, a 7th century BC ruler of Athens.
The rule is: icons that are metaphores for things (like folders or documents) are single-clicked to select and double-clicked to activate; icons that are metaphores for controls (like command buttons or menu items or just about any other widget) are single-clicked to operate the control.
I just hope HP sells of the stuff they make that doesn't suck (calculators, printers, and medical/testing equipment) before they make something really stupid and tank.
They did of course split off test equipment as Agilent Technologies.
Let's ... assume that we *want* network television to continue to be free.
... by watching some commercials targeted at me specifically than paying [money]
I do not accept this premise because it is not true. Network TV is not free. It simply has no cost at the point of delivery. Which, as anyone who has just done their taxes will tell you, is not the same as free.
I'd prefer to skip commercials altogether, but the economic realities mean someone has to pay... I'd much rather pay
You do pay real money for ad-supported TV. You pay through the increased prices of goods and services that are advertised. When you buy a box of soap poweder in the supermarket a large part of the money you hand over goes to pay for all that crap on Fox. The money that pays for ad-supported TV does not magic itself out of thin air - it comes out of your pocket ultimately.
Although TV detector vans are the most obvious way of catching people, the real work is done by computers. They have a bloody big database containing every street address in the UK, and they subtract from this the data from licenses sold. The remainder are households that don't have a license. Since TV ownership is almost universal, almost all of these households will be license evaders.
Corporations need to motivate people to buy their products. ... Capitalism requires product awareness in consumers to work at all.
I buy a new car about once every five years. During prime time TV I am presented with ads for new cars about once every eight minutes. That kind of disparity very quickly becomes annoying.
It all depends how you make your PDFs. The bulk of the size of a typical PDF is in the images, and the images may be embedded within the PDF in various formats and with various levels of compression. A comparison without detailing what format was used is meaningless.
For example, if you scan a typical one page black and white text document at 300dpi and save the result as a TIFF with CCITT Group 4 compression it will be 20-50kB. If you make a PDF by printing to Postscript then using the PDF generation tool within the GhostScript Viewer you will end up with a PDF that is several MB. That's because embedded within the PDF will be an uncompressed bitmap. If you make the PDF using the c42pdf utility your PDF will be the same size as the TIFF, 20-50kB. That's because embedded within the PDF will be the compressed bitmap.
If you use the Adobe Acrobat tools to produce your PDFs, you will find lots of options for controlling how embedded images are compressed, what the resolution is and what formats are used. You can select different options for B&W, greyscale and colour images.
I suspect the DjVu comparison is not comparing like with like. And in any case, as has been noted, anyone who generates PDFs of documents like annual reports with a scanner is missing the point.
I started to get some pain in my right hand from mousing, so I switched to using the left. I was pretty clumsy the first week, a lot better the second, and after that I no longer noticed. Nowadays I switch back and forth every few weeks to even up the wear.
So I would say it is quite possible to use the 'wrong' hand - mousing is a fairly simple motor skill which it seems one can learn pretty fast.
Who says HTML is the answer? If you want an online, threaded, linked version why not refer to the original slashdot articles.
A more appropriate format for a book might be PDF. The Yale Web Style Guide was a reasonable example of how this can work (before they took it away to encourage you to buy the paper copy). Whatever tool Hellmouth was authored with it could almost certainly be rendered as PDF with very little effort.
This is symptomatic of the underlying difference; Unix is written in C and NT is written in C++.
As I remember it, much of the story was actually about how expensive and less-than-ideal Irridium service was going to be due to the political deals that had to be made with third world governments (and their telcos) to get WARC to allocate them the frequencies.
Isn't this the job of Red Hat, et al? Commercial companies who seek to make money from Linux claim that what they bring to the party - that new users can't get from just downloading the sources - is support. Surely, good old market forces will deliver what new users want and need from these companies.